Contour detection from deep patch-level boundary prediction
Teck Wee Chua, Shen Li · 2017
In this paper, we present a novel approach for contour detection with Convolutional Neural Networks. A multi-scale CNN learning framework is designed to automatically learn the most relevant features for contour patch detection. Our method uses patch-level measurements to create contour maps with overlapping patches making interdependent decisions. We show the proposed CNN is very efficient to detect large-scale contours in an image. We further propose a guided filtering method to refine the contour maps by using the extracted large-scale contours. Our contour detection is simple but efficient. Experimental results on the major contour benchmark databases demonstrate the effectiveness of the proposed technique. We show our method can achieve good detection of both fine-scale and large-scale contours.